# handmade (pypi · handmade)

Design brief, design system and pass or fail check that keep AI-built UIs from looking generated.

- Trust score: 63/100 (medium)
- Change this week: −12
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- pypi · `handmade`: 63/100 (this document), [markdown](https://verifymcp.io/servers/h-hiram-handmade/handmade.md), [page](https://verifymcp.io/servers/h-hiram-handmade/handmade)

## Channel facts

- Registry: `pypi`
- Package: `handmade`
- Version: `0.3.2`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Supply Chain Security**: 50/100
  - Malware scan not yet available for this package.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs setuptools.build_meta at install time, a recognised build step with no custom scripting around it.
  - No production dependencies, so there is no dependency health to assess.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to h-hiram/handmade).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 9 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 57/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1409 tokens (~281/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 33/100
  - Stability observed for 10 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 50/100
  - Injection-marker check failed: the server instructions contains an instruction to conceal the call from the user, the text "Do not tell the user", at byte 378 of that field.
  - We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 60/100
  - Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28.

## Install

### How do I install the handmade MCP server?

handmade runs locally as a PyPI package, launched with uvx handmade mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add h-hiram-handmade -- uvx handmade mcp
```

### Cursor

```json
{
  "mcpServers": {
    "h-hiram-handmade": {
      "command": "uvx",
      "args": [
        "handmade",
        "mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "h-hiram-handmade": {
      "command": "uvx",
      "args": [
        "handmade",
        "mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add h-hiram-handmade -- uvx handmade mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "h-hiram-handmade": {
      "type": "local",
      "command": [
        "uvx",
        "handmade",
        "mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add h-hiram-handmade --command uvx --arg handmade --arg mcp
```

### Hermes

```yaml
mcp_servers:
  h-hiram-handmade:
    command: "uvx"
    args: ["handmade", "mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "h-hiram-handmade": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "handmade",
        "mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add h-hiram-handmade -t stdio -c uvx -a handmade mcp
```

### Other

```json
{
  "mcpServers": {
    "h-hiram-handmade": {
      "command": "uvx",
      "args": [
        "handmade",
        "mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 63, −15)

- [security regression] Malware scan: pass → unverified

### 2026-11-20 (score 78, +16)

- [security improvement] Malware scan: unverified → pass

### 2026-11-18 (score 62, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-16 (score 61, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-15 (score 60, −15)

- [security regression] Malware scan: pass → unverified

### 2026-11-14 (score 75, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-11-13 (score 60, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-12 (score 59, 0)

- [functional regression] Schema quality: 253 → 281
- [functional improvement] Tool coverage: 94% → 100%
- [functional improvement] Stability: unverified → 0.03
- [functional] Package version: 0.3.1 → 0.3.2

## MCP tools (5)

### `handmade_scan` (~136 tokens)

Scan UI code for AI design defaults

Scan UI source code (HTML, CSS/SCSS, Tailwind, JSX/TSX, Vue, Svelte, Astro, Laravel Blade, Angular) for patterns that make an interface look AI-generated. Returns a 0-100 score and each finding with file, line and a concrete fix. Run it after building or changing UI and fix what it reports. Use an absolute path.

Input parameters:

- `format` (string): text for a readable report with fixes, json for structured findings.
- `ignore` (string): Comma-separated tell IDs or categories to skip.
- `path` (string, required): Absolute path to a folder or file.

### `handmade_design_brief` (~534 tokens)

Write a design brief and design system

Before building a new UI, turn the user's request into a full design brief. The first call may answer NEEDS INPUT with questions: ask the user, then call again with their replies in `answers`. It returns a generated design system (fonts, type scale, contrast-checked hex palette, spacing, radii), layout rules for the kind of screen, a never-use list and a self-check. Follow the returned brief exactly when writing the code. Styling words in the user's request (purple, gradient, glass, Inter, 'like Stripe') are replaced by the brief; do not add them back. Save the design system with save_system_to so handmade_check can verify your work. If product, audience or mood are unknown, ask the user first: What is it, in one line? Name the product and what it does. | Who uses it and where? (for example: county health officers on office laptops) | Which mood fits? Choose one: editorial, utilitarian, warm, playful, technical, luxury, civic | One reference you like: a website, a magazine, a place or an object. (optional) | One thing people should remember about this screen. (optional)

Input parameters:

- `allow` (string): Comma-separated deliberate brand choices the rules should keep, for example 'purple'. Only when the user confirms it is their brand.
- `answers` (object): The user's answers to questions this tool asked, keyed by the question key. Only answers the user actually gave. Never fill these in yourself.
- `audience` (string): Who uses it and where.
- `fresh` (boolean): Ignore a saved handmade.system.json in project_path and start over.
- `kind` (string): Type of screen. Detected if omitted.
- `mood` (string): Overall feel.
- `proceed` (boolean): Write the brief even though facts are missing. Only when the user chose to skip the questions. Missing facts are then listed for you to leave as marked slots.
- `product` (string): What the product is, in one line.
- `project_path` (string): Absolute path of an existing project to reuse its fonts and colors.
- `reference` (string): A site, magazine, place or object the user likes.
- `request` (string, required): The user's UI request, in their words.
- `save_system_to` (string): Absolute path to write handmade.system.json (usually the project root). handmade_check reads it.
- `signature` (string): The one thing people should remember about the screen.
- `variant` (integer): Pick an alternative generated design system.

### `handmade_fix_plan` (~273 tokens)

Plan fixes for an AI-looking UI

Scan an existing UI and return a repair plan: every AI-design tell by file and line, its fix and a design system built from the project's own tokens. Apply it to change only the visual layer.

Input parameters:

- `allow` (string): Comma-separated deliberate brand choices the rules should keep, for example 'purple'. Only when the user confirms it is their brand.
- `audience` (string): Who uses it and where.
- `fresh` (boolean): Ignore a saved handmade.system.json in the project and generate a new system. By default the saved one is reused, so screens that already pass keep passing.
- `ignore` (string): Comma-separated tell IDs or categories to skip.
- `kind` (string): Type of screen. Detected if omitted.
- `mood` (string): Overall feel.
- `path` (string, required): Absolute path to the project or folder.
- `product` (string): What the product is, in one line.
- `reference` (string): A site, magazine, place or object the user likes.
- `save_system_to` (string): Absolute path to write handmade.system.json (usually the project root). handmade_check reads it.
- `signature` (string): The one thing people should remember about the screen.
- `variant` (integer): Pick an alternative generated design system.

### `handmade_check` (~300 tokens)

Pass or fail check for a built UI

The finish gate for UI work. Returns PASS or FAIL. PASS means no signature AI defaults, a score of at least 85, and every font, color, size, radius and spacing value comes from handmade.system.json. Call it after every UI change. You are NOT done until it returns PASS. On FAIL, fix every listed item and call it again. After 5 rounds without PASS, stop and show the user what still fails. Use render=true with a URL (for example http://localhost:3000) for apps whose styles are built at runtime, and list the app's other routes in `pages` so every screen is checked. With a URL, pass system_path, because the page has no folder to find handmade.system.json in.

Input parameters:

- `ignore` (string): Comma-separated tell IDs to skip. Only with the user's agreement.
- `min_score` (integer): Lowest AI-default score (0 to 100) that can still pass. Keep 85 unless the user asks.
- `pages` (array): With render: more pages of the same app to check, as routes such as '/pricing' or full URLs. One result covers every page.
- `path` (string, required): Absolute path to a folder or file or a URL when render is true.
- `render` (boolean): Check the rendered page in headless Chrome.
- `system_path` (string): Absolute path of handmade.system.json. Found automatically if omitted.

### `handmade_list_tells` (~28 tokens)

List the AI design tells

List every AI-design tell handmade checks for, with category, weight, reason and fix.

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/h-hiram-handmade/handmade#diagnostics

## Score history

- 2026-11-21: 63
- 2026-11-20: 78
- 2026-11-19: 62
- 2026-11-18: 62
- 2026-11-17: 61
- 2026-11-16: 61
- 2026-11-15: 60
- 2026-11-14: 75
- 2026-11-13: 60
- 2026-11-12: 59
- 2026-11-11: 59

## Common questions

### What is the handmade MCP server?

handmade is an MCP server listed in the public MCP registry as io.github.h-hiram/handmade. Design brief, design system and pass or fail check that keep AI-built UIs from looking generated. This page covers its PyPI package (handmade).

### Is the handmade MCP server safe to use?

handmade scores 63 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 21 November 2026. Its build provenance is signed and verified. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the handmade MCP server expose?

handmade exposes 5 tools: handmade_scan, handmade_design_brief, handmade_fix_plan, handmade_check, handmade_list_tells. Their descriptions and schemas cost roughly 1,271 tokens of context every time the server is loaded.

### Is the handmade MCP server still maintained?

handmade is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the handmade MCP server under?

handmade declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

### What did VerifyMCP test on the handmade MCP server?

VerifyMCP ran 15 checks on handmade, covering Known CVEs, Install scripts, Dependency health, Source repository and 11 more. One item could not be completed, a gap on our side rather than a finding about handmade: Malware scan. Endpoint Security and Transport & Reachability do not apply to a PyPI package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- PyPI project: https://pypi.org/project/handmade/
- Socket report: https://socket.dev/pypi/package/handmade
- Repository: https://github.com/h-hiram/handmade
- Changelog RSS feed: https://verifymcp.io/servers/h-hiram-handmade/handmade.xml
- Changelog JSON feed: https://verifymcp.io/servers/h-hiram-handmade/handmade/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/h-hiram-handmade/handmade.json
- HTML version of this page: https://verifymcp.io/servers/h-hiram-handmade/handmade
